Garrad Hassan’s report, “Offshore wind: Economies of scale, engineering resource and load factors”, sought to “provide a supported opinion on the likely load factors for offshore wind power in the UK Round 1 and 2 offshore wind farms”.
Government studies have since cited it as a reference.
The report calculated a range of load factors from 33 to 38 percent, for wind speeds ranging from 8.5 to 9.5 metres per second (See Chart 4).
Actual annual average load factors in fact only passed 35 percent for the first time last year, according to the Department of Energy and Climate Change’s (DECC) “Digest of United Kingdom Energy Statistics 2012”.
For previous years they were in the range 26-30 percent, or 28-35 percent depending on whether they are measured including or excluding turbines under construction.
More recent data show that load factors this year dipped back below the bottom end of the Garrad Hassan range.
A small variation in wind speeds significantly impacts the amount of electricity produced yet speeds have met expectations.
Historical data show that average British wind speeds have averaged around 9 metres per second since 2001, in the centre of the Garrad Hassan projections.
Those observed wind speeds are averaged for the whole of the UK, implying higher values for the superior offshore resource.
Other factors that will impact load factor include: downtime (as a result of technical faults and servicing); wake losses (from turbines caught in the wind “shadow” of others); and electrical transmission losses.
One possible culprit for the lower-than-expected load factors are wake losses.
UK-based renewable energy developer Renewable Energy Systems (RES) estimates that these have been under-estimated as a result of applying onshore wind modelling techniques to offshore farms.
“Research carried out on operational data of offshore wind farms has shown that the actual wake losses for large arrays of turbines are greater than what the industry standard onshore wake models predicted,” the company said in a presentation to an offshore wind conference last year.
RES reported actual wake losses several percentage points higher than as modelled using the “onshore method”.
It estimated actual wake losses for the Horns Rev offshore wind farm off Denmark at 12.4 percent compared with a modelled 8.3 percent, and similar discrepancies for Denmark’s Middelgrunden farm (10 percent versus 6.2 percent) and the Lillgrund farm between Denmark and Sweden (23 percent versus 14.7 percent).
Garrad Hassan’s 2003 report did not explain or reference its estimate for average British offshore wind farm wake losses of 8 percent of output.